Project Description Common disorders of the digestive tract, such as Inflammatory Bowel Disease, colorectal cancer and Barrett's esophagus reflect profound epithelial dysfunction and gene dysregulation. Improved treatments require better understanding of the basis for intestine-specific gene control. Certain transcription factors (TFs) ? CDX2, HNF4 and HNF1 ? act at the distant enhancers of genes that define gut epithelial identity and function, but mechanisms of enhancer assembly, specificity, and activity are not well understood. Nor is it known how chromatin and TF dynamics in gut endoderm allow the emergence of distinctive digestive epithelia. This proposal addresses some of these fundamental questions in mouse intestinal cells in vivo. Among thousands of cis-elements that control intestinal genes, we identify a new class of tissue-specific enhancers that require CDX2 to exclude the repressive histone mark H3K27me3 from large genomic domains; H3K27me3 was not previously implicated at distant enhancers in adult tissues. These findings ascribe a novel, unexpected function for CDX2 at `anti-repressive' Type 1A enhancers that control the most quantitatively vulnerable genes in Cdx2-/- villus cells. Specific Aim 1 addresses the requirements and mechanisms of these novel enhancers. We propose genetic and biochemical experiments to test two central hypotheses: (a) that ~500 key intestinal loci are particularly susceptible to repression by virtue of extensive local deposition of H3K27me3, and (b) that CDX2 recruits the specific H3K27 demethylases KDM6A/B to oppose that locus-wide effect. We will use Eed-/- intestines, which lack methylated H3K27, to determine whether this feature is causally repressive and to attempt rescue of the genes most perturbed in Cdx2-/- intestines. Because Type 1A enhancers show a large difference in H3K27me3 levels between enterocyte progenitors in intestinal crypts and terminally mature enterocytes along the villi, we propose studies to test the hypothesis that this repressive mark limits high expression of enterocyte genes in replicating crypt cells. Specific Aim 2 considers the developmental origins and determinants of intestinal enhancers and asks why certain genomic domains acquire tissue-specific H3K27me3. This Aim builds on our discovery that nascent enhancers for all digestive epithelia are initially accessible throughout the early endoderm and that tissue-specific sites later strengthen in each organ while inappropriate enhancers are shut down. We will test the hypothesis that regions dependent on Type 1A enhancers in the adult intestine are the vestiges of enhancers that were used in development. We will also test the hypotheses that: (a) unless regional TFs reinforce open endodermal chromatin, enhancers shut down in a rostral-caudal wave, and (b) the state of chromatin at any time determines how the developing intestine responds to absence of CDX2. This proposal thus applies innovative state-of-the-art approaches to pursue exciting discoveries and address fundamental questions about intestinal gene regulation in vivo.